key: cord-0253203-ujr7cg98
authors: Pei, Tong-Tong; Kan, Yumin; Wang, Zeng-Hang; Tang, Ming-Xuan; Li, Hao; Yan, Shuangquan; Cui, Yang; Zheng, Hao-Yu; Luo, Han; Dong, Tao G.
title: Breaching the cell-envelope barriers of gram-positive and fungal microbes by a type VI secretion system in Acidovorax citrulli
date: 2021-05-31
journal: bioRxiv
DOI: 10.1101/2021.05.31.446370
sha: bb66f4065443ba31d98f407c02ef997d87eaf462
doc_id: 253203
cord_uid: ujr7cg98

The type VI secretion system (T6SS) is a double-tubular toxin-injection nanomachine widely found in gram-negative human and plant pathogens. The current model depicts that the T6SS spear-like Hcp tube is powered by the contraction of an outer sheath to drill through the envelope of a neighboring cell, achieving cytosol to cytosol delivery. However, gram-positive bacteria seem to be impenetrable to such T6SS action. Here we report that a plant pathogen Acidovorax citrulli (AC) deploys a highly potent T6SS to kill a range of bacteria including Escherichia coli, Pseudomonas aeruginosa, Bacillus subtilis, and Mycobacterium smegmatis as well as fungal species including Candida albicans and Pichia pastoris. Using bioinformatic and biochemical assays, we identified a group of T6SS effectors and characterized one effector RhsB that is critical for interspecies interaction. We report that RhsB contains a conserved YD-repeat domain and a C-terminal nuclease domain. Toxicity of RhsB was neutralized by its downstream immunity proteins through direct interaction. RhsB was cleaved at the C-terminal end and a catalytic mutation within the internal aspartic protease abolished such cleavage. Collectively, the T6SS of AC displays potent activities to penetrate the cell envelope barriers of gram-positive and fungal species, highlighting the greatly expanded capabilities of T6SS in modulating microbiome compositions in complex environments.

ecological impact of the T6SS on diverse natural microbial communities, beyond the previously 1 known susceptible gram-negative and fungal microbes. 2

Next we decided to identify T6SS effectors responsible for the killing activities. The 3 AAC00-1 genome contains 12 vgrG operons that are expected to encode at least one effector per 4 operon (10, 11) . Using bioinformatic and secretome analyses, we identified 17 VgrG-related 5 effector genes and constructed corresponding deletion or inactivating-insertion mutants. Using 6 bacterial competition assays against E. coli and B. subtilis prey, we screened these mutants and 7

found two mutants (Aave_0499, named RhsB and Aave_2838, named RhsE) with impaired but 8 not abolished killing activities (Figure 1 C&D) . These results highlight a functional redundancy 9 of effectors that collectively contribute to killing activities. 10

For the two mutants, both genes encode Rhs-family proteins with an N-terminal PAAR 11 domain, a middle YD-repeat/Rhs domain, and a C-terminal domain of unknown function ( Figure  12 2A). Using Phyre2 sequence analysis, we found no significant hit for Aave_2838 but the C-13 terminal domain of Aave_0499 is distantly related to a virus-type replication-repair nuclease 14 (PDB: 4qbn) with 24% identity (12) . Downstream of rhsB reside two small predicted genes of 15 unknown functions sharing 69% identity and equal length in protein sequence. We name the two 16 downstream genes rimB1 and rimB2 (Rhs-immunity B), respectively. 17

To test if the C-terminus of RhsB (RhsB CT ) is toxic, we expressed it using an arabinose 18 inducible vector in E. coli and compared cell survival in the presence of arabinose (induced) or 19 glucose (repressed). Results show that RhsB CT is highly toxic, reducing survival by 100-fold 20 when induced ( Figure 2B ). Although RhsB CT belongs to the PD-(D/E)XK superfamily (13) , 21 there are no obvious catalytic sites in 4qbn or RhsB CT . We constructed several point mutations 22

around the PD site and tested their toxicity in E. coli. Two mutants, KD-AA and KE-AA, were 23 nontoxic while the K1561A mutant exhibited stronger toxicity than wild type RhsB CT (Figure  1 2B). To test whether RimB proteins confer protection, we constructed the ∆ rhsB-rimB1&2 2 mutant lacking the immunity genes and transformed it with an empty pBBR vector or vectors 3 encoding one or both immunity proteins ( Figure 2C ). Competition analyses against wild type AC 4 or the ∆ tssM mutant revealed that survival was significantly increased when immunity genes 5 were ectopically expressed in the ∆ rhsB-rimB1&2 mutant, suggesting both immunity proteins 6 could confer protection. Using bacterial two-hybrid assays, we found that both RimB1 and 7

RimB2 could bind to the non-toxic RhsB KE-AA construct, while the Aave_0500 protein, encoded 8 downstream of rimB2, did not interact with any of the proteins ( Figure 2D ). As control, Pal and 9

TolB proteins showed positive interaction. 10

Next, we tested whether RhsB is delivered to B. subtilis using a competition assay. The 11 prey B. subtilis was transformed with the empty vector pHT01 or vectors expressing one or both 12 immunity proteins, separately. Results show that all B. subtilis strains expressing the immunity 13 genes survived significantly better than the one expressing the vector alone when they were 14 competed with wild type AC ( Figure 2E ). When the ∆ rhsB mutant was used as the killer, 15 survival of B. subtilis with vector alone was increased to a similar level to that of B. subtilis 16 expressing immunity genes. These results indicate that RhsB is delivered by the T6SS to B. 17 subtilis and its toxicity is neutralized by the RimB immunity proteins expressed in B. subtilis. 18

To determine how RhsB is secreted, we used pull-down analysis and found that RhsB 19 could direct interact with the upstream encoded VgrG3 and conserved chaperone EagT2, 20

suggesting RhsB secretion is mediated by the VgrG spike complex ( Figure 2F ). Deletion of 21 vgrG3 resulted in significantly reduced killing of B. subtilis to a similar level to the rhsB mutant 22

( Figure 2G ). In addition, Western blotting analysis reveals that Hcp secretion was comparable 6 among wild type, the ∆ rhsB, and the ∆ vgrG3 ( Figure 2H ), eliminating the possibility that the 1 reduced killing of ∆ rhsB and ∆ vgrG3 is due to impaired T6SS secretion. Collectively, these 2 results indicate that secretion of RhsB is mediated by VgrG3. 3

To test if RhsB exhibits nuclease activity, we purified the C-terminal domain of RhsB 4 and RhsB KD-AA mutant with an N-terminal 6His-SUMO tag. Results show that DNA was 5 degraded by DNase I and wild type RhsB CT but not by the RhsB KD-AA_CT mutant ( Figure 2I ). 6

Chromosomal inactivation of RhsB also attenuated the killing of B. subtilis to the same level as 7 deletion of rhsB did ( Figure 2J ). In addition, sequence alignment of RhsB with TseI, a self-8 cleavable T6SS effector in Aeromonas dhakensis (14) , shows that RhsB also contains the 9 conserved aspartic catalytic residues for C-terminal cleavage ( Figure 2K ). We thus mutated one 10 of the catalytic residues D1484 to alanine in the non-toxic RhsB KE-AA background. Western 11 blotting analysis shows that the C-terminus of RhsB was also cleaved by the internal protease 12 and the D1484A mutation abolished the cleavage ( Figure 2L ). 13

In conclusion, we report the potent activities of A. citrulli T6SS against a panel of T6SS-based treatment strategies as green alternatives to chemical agents in mitigating infectious 7 diseases in agricultural and medical applications. In short, there seems to be no barrier too thick 8

for the T6SS to break in the microbial world. 9

All strains were routinely grown in LB, 7H9, 7H10 or YPD media following standard culturing 12 conditions for each species. Antibiotics were used at the following concentrations: kanamycin 13 

Cultures were grown aerobically in LB with appropriate antibiotics at 37 °C to OD 600 ~ 2 and 17 collected by centrifugation at 2,500 × g for 8 min. Pellets were resuspended in fresh LB and 18 incubated at 37 °C for 1 h. Cells were centrifuged at 10,000 × g for 2 min twice at room 19 temperature. Pellets were resuspended in SDS-loading dye and used as whole-cell samples. Correspondence and request for materials should be addressed to T. Dong. Survival of prey cells was determined by serial dilutions on selective media. Error bars indicate the mean +/-standard deviation of three biological replicates and statistical significance was calculated using One-way ANOVA test for each group, *P < 0.05, ***P < 0.001, ****P < 0.0001. DL, detection limit. replicates and statistical significance was calculated using One-way ANOVA test for each group, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence

Candida albicans biofilms and human disease

Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system

A view to a kill: the bacterialtype VI secretion system

Assembly and subcellular localization of bacterial type VI secretion systems

The type VI secretion system deploys antifungal effectors against microbial competitors

Translocation of a Vibrio cholerae type VI secretion effector requires bacterial endocytosis by host Cells

A type VI secretion system of Pseudomonas aeruginosa targets a toxin to bacteria

Acidovorax citrulli: Generating basic and applied knowledge to tackle a global threat to the cucurbit industry

The VgrG proteins are "à la Carte" delivery systems for bacterial type VI effectors

Identification of T6SS-dependent effector and immunity proteins by Tn-seq in Vibrio cholerae

FAN1 activity on asymmetric repair intermediates is mediated by an atypical monomeric virus-type replication-repair nuclease domain

SURVEY AND SUMMARY: Sequence, structure and functional diversity of PD-(D/E)XK phosphodiesterase superfamily

Intramolecular chaperone-mediated secretion of an Rhs effector toxin by a type VI secretion system

The bacterial cell envelope

The authors declare no competing interests. 11